Title :
The Wigner-Ville trispectrum: definition and application
Author :
Boashash, Boualem ; Ristich, Branko
Author_Institution :
Signal Processing Res. Centre, Queensland Univ. of Technol.., Brisbane, Qld., Australia
Abstract :
The Wigner-Ville trispectrum (WVT) is defined as a member of a class of time-varying higher-order spectra based on the polynomial Wigner-Ville distributions. It is shown that the WVT is a very efficient time-frequency analysis method for the analysis of FM signals affected by multiplicative noise, and outperforms the Wigner-Ville spectrum (WVS), and therefore any other method based on Cohen´s class of bilinear time-frequency spectra. The authors further consider the instantaneous frequency estimator based on the peak extraction of the WVT. They also develop the statistical properties of such an estimator, for signals affected by both multiplicative and additive Gaussian noise.
Keywords :
random noise; spectral analysis; statistical analysis; time-frequency analysis; FM signals; Wigner-Ville trispectrum; additive Gaussian noise; instantaneous frequency estimator; multiplicative noise; peak extraction; polynomial Wigner-Ville distributions; statistical properties; time-frequency analysis method; time-varying higher-order spectra; Additive noise; Amplitude modulation; Frequency estimation; Frequency modulation; Higher order statistics; Polynomials; Signal analysis; Signal processing; Spectral analysis; Time frequency analysis;
Conference_Titel :
Higher-Order Statistics, 1993., IEEE Signal Processing Workshop on
Conference_Location :
South Lake Tahoe, CA, USA
Print_ISBN :
0-7803-1238-4
DOI :
10.1109/HOST.1993.264555